PROJECT SUMMARY/ABSTRACT The recent development of methods that allow continuous laboratory evolution of biomolecules has made it increasingly possible to generate proteins with new, tailored activities for next-generation therapeutics. In particular, phage-assisted continuous evolution (PACE), a method that allows proteins to undergo directed evolution at a rate of ~100-fold faster than conventional methods, has recently been used to evolve new activities in a number of proteins, including RNA polymerases, Cas9 proteins, and viral proteases. While these early applications illustrate the potential of the PACE system, there remain intrinsic technical barriers that limit the success rate, efficiency, and wider application of PACE for creating highly selective, designer molecular therapeutics. The first barrier is the exceedingly low throughput with which PACE experiments can be conducted in parallel, which greatly limits the number of evolutionary trajectories that can be assessed and prohibits large-scale evolution of variants with diverse specificities/activities. The second is an inability to precisely and dynamically control PACE selection conditions (positive and negative), which is critical for fine- tuning properties such as the selectivity of evolved proteins and for achieving successful PACE outcomes. We propose to overcome these barriers by developing an automated, high-throughput system for PACE with individual, real-time monitoring and control over selection conditions (ePACE). To accomplish this goal, we will adapt eVOLVER, a scalable do-it-yourself (DIY) framework we recently invented that uniquely enables scaling both throughput (>100 vials) and individual programmable control of culture conditions during continuous cell growth. Leveraging the highly modular and open source wetware, hardware, and web-based software of eVOLVER will allow us to develop ePACE with a projected throughput ~50-100-fold greater than current PACE technology, with setup costs of >10-fold lower, and the capability of programming real-time, algorithmically- driven modulation of selection conditions to comprehensively explore directed evolution landscapes. We will then demonstrate the ePACE system in two directed evolution case studies that specifically highlight and test the benefits of our enhanced functionalities. The first study will apply the high-throughput capabilities of ePACE to perform multiplex evolution of Cas9 (CRISPR) variants with compatibility for every possible PAM sequence, a large scale evolution that is impractical for traditional PACE. In the second study, we will apply adaptive (closed-loop) selection stringency modulation to the traditionally challenging problem of reprogramming proteases toward new, intracellular therapeutic targets. This effort will seek to acquire a Botulinum neurotoxin protease variant capable of selectively cleaving caspase-1, toward an ultimate goal of a deliverable, caspase- activing protease for potential cancer therapies. This work will provide a standardized, democratic, and powerful platform to streamline and expand the scope of directed evolution methods for rapidly creating new molecular entities and therapeutics.